After basic IS-IS functions are configured, you can build a Layer 3 network using the IS-IS protocol. If other IS-IS functions are required, configure them according to reference sections.
Table 1 describes the IS-IS configuration tasks.
Scenario |
Description |
Task |
---|---|---|
Configuring basic IS-IS functions |
To deploy the IS-IS protocol on IPv4 networks, configure basic IS-IS functions to enable communication between different nodes on the network. Other IS-IS features can only be configured after the basic functions are configured. |
|
Configuring IS-IS network security |
On IS-IS networks, unauthorized users can attack the IS-IS network by modifying data packets or forging authorized users. To ensure security of services carried on IS-IS networks, configure area or domain authentication and interface authentication. |
|
Configuring IS-IS route selection |
If multiple redundant links are available in the network using the IS-IS protocol, the route in the IS-IS routing table may not be the expected optimal route. This does not meet the network planning and traffic management requirements. To optimize the IS-IS network and facilitate traffic management, more accurate control of the routes on the network is required. |
|
Configuring IS-IS routing information exchange |
In practical applications, to meet network requirements, configure route policies to accurately control advertising and receiving of IS-IS routing information. |
|
Configuring IS-IS route aggregation |
Route aggregation allows multiple routes with the same IP prefix to be aggregated into one route. Route aggregation on a large IS-IS network can effectively reduce entries in the routing table. This minimizes system resource consumption and facilitates management. In addition, if a link in the aggregated IP address segment frequently alternates between Up and Down, devices outside this segment will not be affected by the change. This prevents route flapping and improves network stability. |
|
Configuring IS-IS route convergence |
To enable IS-IS to rapidly detect the network changes, speed up the IS-IS network convergence. To minimize the effect on networks from route flapping and reduce load on the device, slow down the IS-IS network convergence. |
|
Configuring LSP fragment extension |
When information contained in the LSP data packet Protocol Data Unit (PDU) to be advertised by IS-IS increases greatly, the IS-IS device will generate multiple LSP fragments to carry and advertise more information. |
|
Configuring mesh groups |
On the NBMA network, when an interface of the switch receives a new LSP, the LSP is flooded to other interfaces of the switch. On highly-connected networks that have multiple P2P links, this processing method results in repeated LSP flooding and wastes bandwidth resources. To solve this problem, create a mesh group and add some interfaces to the group. The switch never floods the LSPs received at interfaces in the mesh group to other interfaces from the same group, and only floods the LSPs to interfaces from other groups or interfaces that are not configured to any mesh groups. |
|
Improving IS-IS reliability |
To ensure rapid recovery from failures on networks using the IS-IS protocol, adopt the solution of fast fault detection and standby link switchover. However, the IS-IS fault detection mechanism and link switchover require a long period of time, which fails to meet the requirements of services that are highly sensitive to packet loss and packet delay. To meet requirements of these services, use BFD for IS-IS to implement fast fault detection and use IS-IS GR helper to implement fast switchover. This improves IS-IS reliability. |
|
Configuring IS-IS overload |
If the system cannot store new LSPs or synchronize the LSDB normally, the calculated routing information will be incorrect. In this case, the system can enter the overload state. Routes reached through the device will not be calculated, but routes directly connected to the device will not be ignored. When an IS-IS device on the network requires upgrade or maintenance, the device needs to be temporarily isolated from the network. To prevent other devices from forwarding traffic through this node, set the overload bit for the device in question. |